By Robert D. Owen1
In recent years, swiftly evolving technology has produced ever more exciting
and useful ways to generate electronically stored information (ESI), and data
volumes have quietly continued their humorless climb.
Judicial officers have struggled to apply our nation's 1938 philosophy of
full pretrial disclosure to this new reality, but their well-intentioned efforts
have produced inconsistent results, especially in the areas of preservation and
The inconsistencies have derived from differing circuit court culpability
requirements, differing facts in the cases presented for decision, but also,
unfortunately, from inherent differences in some individual judges'
predilections towards corporations.
For corporate and government litigants exposed to constant litigation, the
inconsistencies are resolved by a “race to the bottom,” leading to adherence to
the strictest and most costly interpretation of their preservation duties--a
coping mechanism that is hugely wasteful in a time of contracting resources and,
to use a fashionable but grimly accurate word, unsustainable.
At stake is whether the civil justice system, like the American dollar
threatened by a ballooning national debt, will survive the future.
Fortunately, contra the Article I and II branches of our government, the
Article III branch does appear to have the necessary leadership and political
will to take some positive steps. As the nation edged closer to the eDiscovery
cliff a few years ago, alarms were sounded and heeded, and a robust national
discussion of what to do ensued.
The first significant event came in the fall of 2009, when the American
College of Trial Lawyers and the Institute for the Advancement of the American
Legal System co-published a survey and report that highlighted the system's
problems, especially concerning eDiscovery.
The Civil Rules Advisory Committee followed with the Duke Litigation Review
Conference in May 2010 (which had been preceded by a “dress rehearsal” meeting
organized by the Sedona Conference, which I co-chaired with William P.
Butterfield, partner at Hausfeld LLP).
In September 2011, the Dallas Mini-Conference focused the Rules Committee
members and select attendees on the vexing questions of preservation and
sanctions. Many approaches to amending the Federal Rules of Civil Procedure were
discussed and debated, and the dialogue has continued in the 16 months since
Now, thanks to the leadership of District Judge David G. Campbell (D. Ariz.)
and District Judge (former Magistrate Judge) Paul W. Grimm (D. Md.), proposed
amendments have emerged which many believe will help pull us back from going
over the eDiscovery cliff.
One concerns preservation, specifically the current Rule 37(e); the other
deals with the scope of discovery, Rule 26(b)(1), which defines and guides the
outer limits of preservation.
Today's affirmative duty to preserve evolved out of the ancient negative
prohibition against destroying evidence relevant to a lawsuit--so in a very real
sense, current decisions requiring companies to freeze virtual mountains of
possibly relevant materials are the descendants of centuries-old decisions
making it wrongful, for example, to discard a broken axle at the center of an
There is currently no civil rule--in federal or in state courts--that defines
a prospective litigant's duty to preserve evidence. Individual judges in
individual cases have to base their decisions on the amorphous “inherent power”
concept and the “common law.” Not surprisingly they have come to different
conclusions on certain aspects of the preservation duty, producing the
inconsistencies mentioned above.
For example, while some judges hold that a written litigation hold must be
issued in every case, others have held that other routes suffice provided the
path taken is reasonable.
Some have held that mere negligence in preservation can justify sanctioning a
party, while others have required at least recklessness or more intentional
Exactly when the preservation duty is triggered has been a particularly
difficult question, so some companies play it safe by over-preserving, to avoid
being sanctioned for having spoliated evidence. Microsoft, for example, reported
to the Rules Committee at the Dallas Mini-Conference that two-thirds of its
14,000 litigation holds pertained to unfiled disputes.
In response, the Committee has crafted a new Rule 37(e), which if adopted
will impose a uniform national standard as to when sanctions can be issued for a
failure to preserve.
New Rule 37(e) would work hand in glove with a new Rule 26(b)(1) that would
narrow the scope of discoverable material in significant ways and insert
proportionality into the text of the rule, thereby reinforcing the common sense
limitations that new Rule 37(e) would apply to the preservation duty. I will
discuss the possible change to Rule 26(b)(1) in a future article.
The current version of Rule 37(e) was born as Rule 37(f) in the 2006
amendments, where was known (somewhat inaccurately) as the “safe harbor”
provision. (The textual recasting of the rules since 2006 resulted in the
section being renumbered.)
It was an attempt to protect producing parties--usually corporations, but not
always--from knee-jerk sanctions if their “auto-delete” functions purged data
that was later deemed discoverable, provided that the purging occurred during
the “routine, good-faith operation” of their systems. It was narrowly focused on
what the comments called a “distinctive feature of computer operations, the
routine alteration and deletion of information that attends ordinary use” and it
only applied to electronic information managed in “good faith”-- in other words,
without an intent to use the feature to deny the information to the requesting
party. It did not provide broad guidance concerning preservation generally nor
define when sanctions or remedies were justified.
Unfortunately, the Committee Note put an unnecessary gloss on the term “good
faith,” saying that it “may involve a party's intervention to modify or
suspend certain features of that routine operation to prevent the loss of
information, if that information is subject to a preservation obligation.”
Some decisions soon construed “may” to be “must” and so in many cases the
protections of the rule were unavailable to parties who did not institute a
legal hold andsuspend auto-delete as soon as litigation was “reasonably
The safe harbor became in effect a much narrower safe cove, and
although some cases “got it right,” enough did not so that the rule never really
achieved the goal of predictable, definitive protection.
Litigants still play “gotcha” over preservation issues, still create
spoliation sideshows, and still exploit the fear of a sanction order to drive
settlements whose value is often untethered to the underlying merits.
The Rules Committee and the Judicial Conference Standing Committee have
recently approved a new version of Rule 37(e) for publication and comment.
(Publication will likely occur in August 2013, assuming that no major changes
The current version of 37(e) would be replaced with a new section entitled
“Failure to Preserve Discoverable Information.” As “restyled” by advisors to the
Standing Committee, which approved it at their January 2013 meeting, the text
that appears in the box on the following page represents the current form of the
(e) FAILURE TO PRESERVE DISCOVERABLE INFORMATION.
(1) Curative measures; sanctions. If a party failed to preserve
discoverable information that reasonably should have been preserved in the
anticipation or conduct of litigation, the court may
(A) permit additional discovery, order the party to undertake
curative measures, or order the party to pay the reasonable expenses, including
attorney's fees, caused by the failure; and
(B) impose any sanction listed in Rule 37(b)(2)(A) or give an
adverse-inference jury instruction, but only if the court finds that the
(i) caused substantial prejudice in the litigation and was
willful or in bad faith; or
(ii) irreparably deprived a party of any meaningful opportunity
to present a claim or defense.
(2) Determining reasonableness and willfulness or bad faith. In
determining whether a party failed to preserve discoverable information that
reasonably should have been preserved, and whether the failure was willful or in
bad faith, the court should consider all relevant factors, including:
(A) the extent to which the party was on notice that litigation was
likely and that the information would be discoverable;
(B) the reasonableness of the party's efforts to preserve the
(C) whether the party received a request to preserve information,
whether the request was clear and reasonable, and whether the person who made it
and the party engaged in good-faith consultation about the scope of
(D) the proportionality of the preservation efforts to any
anticipated or ongoing litigation; and
(E) whether the party timely sought the court's guidance on any
unresolved disputes about preserving discoverable
There is much to applaud in this draft, but its most striking feature is its
requirement that before issuing sanctions a district court must find that the
failure to preserve caused “substantial prejudice” and was “willful or in
The intended effect is to eliminate the differences among circuits and
districts that currently exist on two important issues: (i) the mental state of
the party that failed to preserve certain discoverable material, and (ii) the
amount of prejudice that a movant seeking sanctions must show.
Companies have been fearful for years that an innocent mistake despite best
intentions could result in a sanction order. Over-preservation was the logical
result. As noted earlier, Microsoft's submission to the Rules Committee in 2011
gave concrete examples, reporting that for every “one-page trial exhibit,
Microsoft … preserves almost 340,000 pages.”
The new Rule, combined with the narrowing of the scope of discovery that may
be adopted in a revised Rule 26(b)(1)-- stay tuned for Part II of this
article--could materially reduce this senseless waste.
Unfortunately, the drafters approved an exception to that requirement of
showing that a failure to preserve caused substantial prejudice and was willful
or in bad faith. Section (e)(1)(B)(ii) is included to allow a sanction where,
even though there may not have been willfulness or bad faith, a party was
“irreparably deprived” “of any meaningful opportunity to present a claim or
defense” by a failure to preserve.
If there was no fault, why should there be a sanction? Isn't this an
evidentiary issue, rather than an eDiscovery one? This exception, although
intended by the drafters to be narrow, opens up the rule to dilution by case
law, in the same way the current version of 37(e) was.
The new Rule would authorize issuance of any of the 37(b)(2)(A) sanctions
without a showing that a prior court order had been disobeyed. Although common
law decisions based on inherent power have imposed similar sanctions for years
as punishment for spoliation, the current Federal Rules expressly allow such
sanctions in only one instance absent violation of a court order. (Rule for
16(f)(1)(B) which authorizes imposition of such sanctions if a party is
unprepared or fails to cooperate in a 16(b) conference.)
It should also be noted that one of the express goals of the Rules Committee
in preparing this draft was to discard reliance on the courts' inherent
authority. This is a welcome development. If adopted along with the Draft
Committee Notes proposed, the Rule would provide much clearer guidance to courts
and parties as to where the limits are, although a major shift in decisional
results is not expected.
The Committee notes that its study of inherent authority decisions revealed
that all were consistent with the “substantial prejudice and bad faith or
willfulness” requirements included in proposed 37(e)(1)(B)(i).
Another major feature of the Rule is the express distinction it makes between
“curative measures” and sanctions, which serves as a helpful reminder that a
district court has many responses short of sanctions to which it can resort
where material has been lost.
A sanction order sticks forever, and one sanction makes future adverse
results more likely (think, e.g., Voom v. Echostar), which is why
corporations with their reputations and future litigation postures at stake are
so fearful of such an order being entered in even a single case.
The same is true of a sanctioned lawyer: the history of such an order will
follow the lawyer in perpetuity and tarnish his pro hac vice applications, and
perhaps even his career (think, e.g., the sanctioned lawyers in the
Qualcomm case, some of whom continue to suffer career prejudice even
after the sanctions were lifted).
Subsection (2) of the proposed rule comprises a list of five factors that
provide guidance for courts facing decisions as to whether a party acted
reasonably, willfully or in bad faith for purposes of subsection (1).
The reasonableness factor--long a legal touchstone in this area--appears in
factor (B), even though arguably its centrality to eDiscovery case law suggests
it should have been positioned as factor (A).
First place in the list is instead given to whether the party was on notice
“that litigation was likely and that the [lost] information would be
discoverable” in that litigation. This, of course, is a restatement of the
“anticipation of litigation” trigger, which is also expressed in the first
paragraph of the draft Rule 37(e) as follows:
“If a party failed to preserve
discoverable information that reasonably should have been preserved in the
anticipation or conduct of litigation, …”
Thus, because the rule makers are hobbled by the Rules Enabling Act from
making rules that apply before the commencement of an action, they have instead
with this rule authorized district courts, once a matter is on file, to “look
back” and assess what a party did before commencement but after it was
“on notice that litigation was likely.”
This unfortunately preserves--and fails to clarify--the ambiguous “reasonable
anticipation of litigation” trigger with all its attendant uncertainties and
constitutes, in my view, the weakest aspect of an otherwise good proposal. In
2011, I argued to the Rules Committee at the Dallas Mini-Conference and in a
letter re-published in DDEE that the rights of all parties would be
reasonably balanced, and the uncertainty of the "anticipation of litigation"
trigger would be eliminated, by a rule requiring preservation only upon
commencement of a federal action, if coupled with a prohibition against
intentional destruction of material in order to make it unavailable to others in
future litigation (See, 11 DDEE 451, 11/10/11).
On the question of notice, the drafters included factor (C) which focuses on
whether the party received a preservation demand notice. This factor was the
subject of some disagreement among Rules Committee members, who adopted it on a
7-4 split vote of the committee.
Some members feared that by giving legal acknowledgement to such notices--a
first in the rules--the provision might encourage their widespread use and
abuse. Other members favored the clause as a way of promoting communication
Another controversial factor was rejected. The district court would have been
authorized to weigh a party's “resources” and “sophistication in litigation”
when considering imposing sanctions. That provision was rejected by the
committee on a 6-5 vote. Opponents argued that such a factor would disadvantage
corporations and the government alike.
Factor (D) expressly inserts proportionality into the analysis, and would
allow the court to consider whether the preservation efforts undertaken by the
party that lost data were proportional to the case. This constitutes important
recognition that mammoth resources--even if available to a corporation or the
government--need not be devoted to preserving every last bit of data in a
smaller or less consequential matter. However, inserting proportionality into
Rule 26(b)(1), which is under active consideration by the Rules Committee, would
help to achieve the same goal.
Finally, factor (E) encourages a preserving party to seek the court's
guidance on preservation disputes. Of course, pre-filing there is no court from
which a party can seek guidance, which is a problem that plagues
well-intentioned potential defendants bound to start preservation upon only the
anticipation of litigation.
A principal concern is whether the five factors listed in (e)(2)(A)-(E) are
necessary at all. Each has the potential of detracting from the principal focus
of the amendment, which is to require a showing of substantial prejudice and bad
faith before sanctions can be issued. A straightforward definition of “bad
faith” seems preferable to the diffusion that could result over time from
including the five factors, e.g., “intentional conduct designed to deprive the
requesting party of the information that was not preserved.”
One open question will be whether the failure to suspend auto-delete is
sufficiently “willful” to justify sanctions. (Since “willful” actions can be
taken in good faith, it would seem preferable simply to drop the willfulness
requirement from (e)(1)(B)(i) altogether.)
Another key question will involve the linkage, if any, between substantial
prejudice and “materiality.”
Finally, hanging over all these questions is the issue of how the
“irreparably deprived” exception in 37(e)(1)(B)(ii) will be construed, and
whether it opens so much of an exception to the bad faith and willfulness
requirements that it will make them irrelevant. (Notably on the latter point,
the exception was subject of some criticism at the January 3 and 4 Standing
Committee meeting, which reportedly asked that the Rules Committee restudy it at
its April meeting.)
The draft Rule fails to require that any sanction imposed be proportional to
the importance of the underlying claim, which would seem advisable. Nor does it
make any provision for Act of God destructions of information that “irreparably
deprive” a party of a meaningful opportunity to present a cognizable claim or
defense, and neither party is at fault.
But in the end this is a very positive proposal that would create some
national standards and overrule some of the scarier spoliation decisions. The
public comment period will provide an opportunity to improve this draft. Thanks
to all who have contributed to this effort.
In my next article, I will report on the status of the proposed amendment to
Rule 26(b)(1). As yet that proposal has not been reported out of the Rules
Bob Owen is a partner in the New York office of Sutherland Asbill &
Brennan LLP. He has more than 35 years of experience in commercial litigation in
New York City and around the country. A nationally recognized adviser to
financial services, energy, and technology companies, Bob is known for creative
and efficient dispute resolution. Bob is widely recognized for his leadership in
eDiscovery, and recently became President of the not-for-profit Electronic
Discovery Institute, and a member of the Advisory Board of the Georgetown Law
Advanced eDiscovery Institute. He is also a member of the editorial advisory
board of Law Technology News, a participant in The Sedona Conference® Working
Group 1, and is listed by Chambers Global and Chambers USA (Band 1) in the
eDiscovery field. He is also a licensed private
1 The author wishes to thank Thomas Y.
Allman, who generously reviewed this article in draft and made suggestions that
substantially improved it, including the suggested definition of “bad faith”
contained in the “Reservations and Conclusions” section.